Real-world time-series data often suffer from missing observations, hindering long-range temporal modeling. However, most existing imputation methods formulate imputation as conditional reconstruction over limited context, which restricts temporal information propagation and fails to explicitly model temporal evolution. To overcome this limitation, we propose the Conditional Temporal Inference Paradigm (CTIP), which formulates time-series imputation as conditional inference along temporal evolution. Under this paradigm, we introduce CBiT, which leverages a history compression mechanism to encode long-range history into a compact latent space for history-conditioned temporal imputation. In addition, we adopt a partitioned modeling strategy that distinguishes historical context and temporal imputation targets with only linear-time complexity. Extensive experiments on multiple public benchmarks show that CBiT improves imputation accuracy by reducing Masked MAE and Masked RMSE by 27.3% and 18.6%, respectively, across different missing rates.

The optimal path problem is an important topic in the current geographic information system (GIS) and computer science fields. The Dijkstra algorithm is a commonly used method to find the shortest path, which is usually used to find the least cost path from a single source. Based on the analysis and research of the traditional Dijkstra algorithm, this paper points out the problems of the Dijkstra algorithm and optimizes it to improve its storage capacity and operation efficiency. Then, combined with the traffic elements, a new network-based optimal path planning method is established. However, the existing network is far from actual operation in terms of the expansion of the transportation network, the uncertainty of the transportation environment, and the differences in the transportation area. Therefore, this paper proposes an optimal transshipment path planning method based on deep learning, which is oriented to multimodal transportation scenarios. This paper mainly introduces the intelligent transportation system and intelligent navigation system, and then conducts in-depth research on optimal path planning. This paper also uses the deep neural network algorithm to optimize the calculation, and finally analyzes its use and application. Simulation experiments were also performed to analyze the relationship between energy consumption, emissions, speed, load cost, and other factors under the optimal path. The final experimental results show that within the range of the emission limit of [100,200], the emission is 50%, the emission is less than 100%, but the emission is higher than 75%. In [100,200], 75% of the loading rate emits no less than 100%. In [200,300], the 50% and 100% emissions are the same. This also means that the emissions are the same but the paths are not necessarily the same.

在多因素耦合作用下桥梁结构会产生振动,以毫米波雷达技术为测量手段,提出桥梁索杆振动检测及自动控制方法.分析在恒载作用下桥梁索杆静止状态与振动状态的应变能,结合动能、势能和恒载做功等建立索杆平面内与平面外的振动特征描述方程.通过发送多周期连续毫米波雷达探测信号,取得同周期回波信号.根据回波信号与对应瞬时相位得到单频差拍信号及瞬时相位,检测出索杆振动幅值.利用半正定矩阵和正定矩阵等函数,建立令振动幅值达到最小值的桥梁索杆振动自动控制模型.实验结果表明,所提方法检测精准度较高,自动控制效果较好,并有效抑制了振动幅度,具备良好的应用性能.

In the current integrated circuit (IC) design process, research on its design process is relatively backward. This paper mainly adopts 5G + artificial intelligence technology, and conducts an in-depth discussion on the structure and algorithm of wireless network integrated circuits. The article analyzes the development of IC industry and the application of EDA software, and puts forward some problems that China’s IC industry is currently facing. The article discusses the realization of IC design flow, which is realized from five aspects: design input, function simulation, layout realization, physical verification, and post-parasitic simulation. Then, in view of the problems existing in the current IC design process, specific solutions for improvement and strengthening are given. Four parts are designed in detail, namely multi-mode simulation, realization of fully automated layout, feature extraction and modeling, IR Drop and EM analysis. Combined with the widely used characteristics of IP design reuse technology, this paper conducts a targeted research on the process of IP design reuse. From the IP circuit transplantation and IP layout transplantation, the specific discussion is carried out. From the perspective of IP design reuse, the design process of the wireless network integrated circuit is further improved. This paper turns to the design of RF IC from the perspective of IC design process, giving the flow chart of RF chip. The test results showed that over the full temperature range, the total current was about 2.45 mA, which met the design requirements. When the load resistance is 4KΩ, the ascending propagation delay is 92tns and the descending propagation delay is 65tns, both of which increase significantly. The article can provide further reference for the design of wireless network integrated circuits.

The emergence of fault prediction and health management (PHM) technology has proposed a new solution and is suitable for implementing the functions of improving the intelligent management and control system. However, the research and application of the PHM model in the intelligent management and control system of electronic equipment are few at present, and there are many problems that need to be solved urgently in PHM technology itself. In order to solve such problems, this paper studies the application of the equipment-status-assessment method based on deep learning in PHM scenarios, in order to conduct in-depth research on the intelligent control system of electronic equipment. The experimental results in this paper show that the change in unimproved deep learning is very subtle before the performance change point, while improvements in deep learning increase the health value by about 10 times. Thus, improved deep learning amplifies subtle changes in health early in degradation and slows down mutations in health late at performance failure points. At the same time, comparing health-index-evaluation indicators, it can be concluded that although the monotonicity of the health index is low, its robustness and correlation are significantly improved. Additionally, it is very close to 1, making the health index curve more in line with traditional cognition and convenient for application. Therefore, an in-depth study of methods for health assessment by improving deep learning is of practical significance.

Wireless sensor network is a network that integrates sensor technology, computer technology, information processing technology, and communication technology. This paper aims to study how to analyze and study the routing optimization of wireless sensor network based on deep learning and describe the neural network. This paper puts forward the problem of routing optimization, which is based on the dynamic programming of wireless sensor network, and then elaborates around its concept and related algorithms and designs and analyzes the case of wireless sensor network optimization. Through the comparative analysis of the five algorithms in computer simulation, although the average network delay performance of DPER reached 0.47 s, it could effectively prolong the life cycle of the network. The DPER algorithm not only improves the network life but also improves the network energy utilization rate, shortens the average path length of the network, and reduces the standard deviation of the remaining energy of the node.

The monitoring of bridge dynamic displacement under normal operation conditions has been a vital need for the assessment of the serviceability of bridges. Traditionally, it was mostly accomplished by the Global Navigation Satellite System (GNSS). However, the poor measurement accuracy and low sampling rate of the GNSS limit the use of monitored displacement signals. In this paper, we propose a novel adaptive multirate Rauch-Tung-Striebel (RTS) smoother that fuses the measurement signals of GNSS and accelerometers to improve both accuracy and sampling rate. The proposed algorithm distinguishes itself from previous studies by adaptively estimating the unknown time-varying transition and GNSS measurement noise variances using the variational Bayes (VB) technique, making it more accurate and less human-involved. The proposed algorithm was validated on a field test conducted on Chishuihe Hongjun Bridge in China, which has the second longest main span and the second highest main towers among suspension bridges over valleys in the world. The algorithm estimated the dynamic displacement at an accuracy of 2.09 mm, which is 21.4% better than the result of the previous algorithm, with both pseudo-static and several low-order main vibration mode components recovered.

The recent development of Internet-of-Things (IoT) technologies has enabled smaller and lower-cost sensor nodes, motivating the deployment of more flexible and scalable sensor networks for infrastructure monitoring applications. However, because these nodes tend to be affected by environmental conditions and aging, they are prone to long-term drift over years of operation; thus, they need to be recalibrated on a regular basis to ensure data accuracy. In this paper, we propose an in situ blind calibration algorithm for infrastructure monitoring sensor networks, which requires neither physical intervention nor the assumption that the sensors are measuring identical ground-truth signals. The algorithm uses a multioutput Gaussian process (MOGP) to model the spatial-temporal distribution of the measurand and drift, thus removing irrelevant short-term fluctuations and decomposing the drift from long-term trends. We evaluate the algorithm on a real-world dataset, and the results show that the proposed method can successfully calibrate a number of drifting sensors more than 20% greater than previous methods while achieving a higher drift estimation accuracy.

近年来,我国的桥梁建设的发展迅速,桥梁作为承载交通的重要构筑物.随着我国公路桥梁建设技术的发展和日益完善的公路系统,桥梁建设技术也得到长足的进步,相继建成了各类桥型的世界级跨径桥梁.但是,桥梁的管养水平却仍处于初级阶段.从二十世纪八十年代起,我国许多特大型桥梁陆续安装了健康监测系统,对维护桥梁安全起到了一定作用,但由于技术、系统等多方面原因,始终未发挥出最大的作用.经调查研究和近年来工程实际验证,对于桥梁结构安全状态的评估,不仅需要结构的各种外荷载作用下的静动力响应的分析数据,还需要对桥梁构件及其附属设施的外观检查、外部环境的记录,对检测、监测、检查的所有数据和视频、图像甄别分析,而这些是早期从国外引进健康监测系统无法达到的.

无线智能传感器结合云平台可以实现建筑结构的长期健康监测,模态识别是结构健康监测的重要内容.希尔伯特黄变换(Hilbert-Huang Transform,HHT)因其适用于非线性非平稳信号,且具有完全自适应性等特点,在模态识别领域中被广泛应用.与实验室中进行结构模态参数识别不同的是,长期监测中模态参数识别的算法不能出现主观的参数选择过程,而传统HHT的第一步经验模态分解(Empirical Mode Decomposition,EMD)会产生虚假的固有模态函数(Intrinsic Mode Function,IMF)分量,对虚假分量的识别与剔除往往依赖研究人员的主观判断.该文提出了一种基于深度神经网络(Deep Neural Networks,DNN)与K-L散度(Kullback-Leibler Divergence,K-L Divergence)的新算法,可以自动化识别并剔除EMD产生的虚假分量,从而使得EMD后得到的固有模态函数均为真实分量.

隧道实时监测系统是确保隧道正常、安全运营的关键因素之一,但传统的隧道监测系统存在数据繁多、可读性差、紧急情况下难以评估等缺点.基于此,提出了新型的设计理念,即将物联网技术的运维周期结构安全监测系统与建筑信息模型技术相融合,研发了可以实现隧道安全运营的多用户协同管理以及监测与养护一体化运维管理的隧道实时监测系统,并以某隧道为试点,验证了建筑信息模型技术在隧道监测中的应用效果.

In modern wireless sensor network (WSN) applications, the long-term drift of sensors is becoming a challenge for the accuracy and reliability of the data. In this paper, we propose a blind calibration algorithm for WSNs. The algorithm models the spatial and temporal correlation with multi-output Gaussian process (MOGP) from a long-term perspective. It can estimate the drift with mean square error (MSE) less than 10% on two real-world datasets, which outperforms other novel blind calibration algorithms, especially when the signal contains long-term trend.

Recently, Deep Learning (DL), especially Convolutional Neural Network (CNN), develops rapidly and is applied to many tasks, such as image classification, face recognition, image segmentation, and human detection. Due to its superior performance, DL-based models have a wide range of application in many areas, some of which are extremely safety-critical, e.g. intelligent surveillance and autonomous driving. Due to the latency and privacy problem of cloud computing, embedded accelerators are popular in these safety-critical areas. However, the robustness of the embedded DL system might be harmed by inserting hardware/software Trojans into the accelerator and the neural network model, since the accelerator and deploy tool (or neural network model) are usually provided by third-party companies. Fortunately, inserting hardware Trojans can only achieve inflexible attack, which means that hardware Trojans can easily break down the whole system or exchange two outputs, but can't make CNN recognize unknown pictures as targets. Though inserting software Trojans has more freedom of attack, it often requires tampering input images, which is not easy for attackers. So, in this paper, we propose a hardware-software collaborative attack framework to inject hidden neural network Trojans, which works as a back-door without requiring manipulating input images and is flexible for different scenarios. We test our attack framework for image classification and face recognition tasks, and get attack success rate of 92.6% and 100% on CIFAR10 and YouTube Faces, respectively, while keeping almost the same accuracy as the unattacked model in the normal mode. In addition, we show a specific attack scenario in which a face recognition system is attacked and gives a specific wrong answer.

A wireless smart sensor is designed for cable tension monitoring system of bridges.The smart sensor possesses several technical features such as ultra-low power consumption,built-in fundamental frequency extraction algorithm and capability of wireless communication through the self-organized wireless sensor network.The stability and accuracy of our fundamental frequency extraction algorithm are verified in both experiments and long-term cable-stayed bridge monitoring applications.Thank to the welldesigned hardware and software,the battery-powered sensor is able to keep on working for more than ten years.The wireless smart sensor proposed in this paper can largely cut down the complexity of the cable tension monitoring system construction and reduce the difficulty of data analysis.At the same time,the cost of maintenance can also be greatly depressed.

本文针对索力的应用场景,设计了一款智能无线索力传感器.该传感器具有超低功耗、前端智能算法、无线自组织网络等技术特点.提出并实现了在超低功耗嵌入式硬件平台上的索频提取、振幅提取方法,经实验及实际工程检验,该方法测量结果准确.该设计整机平均功耗不超过100uA@3V,可在电池供电环境下实现常年连续工作.此设计有效地解决了拉索监测过程中监测系统建设冗繁、数据分析难度大、系统成本偏高、维护复杂等实际工程问题.

Temporal drift of sensory data is a severe problem impacting the data quality of wireless sensor networks (WSNs). With the proliferation of large-scale and long-term WSNs, it is becoming more important to calibrate sensors when the ground truth is unavailable. This problem is called ”blind calibration”. In this paper, we propose a novel deep learning method named projection-recovery network (PRNet) to blindly calibrate sensor measurements online. The PRNet first projects the drifted data to a feature space, and uses a powerful deep convolutional neural network to recover the estimated drift-free measurements. We deploy a 24-sensor testbed and provide comprehensive empirical evidence showing that the proposed method significantly improves the sensing accuracy and drifted sensor detection. Compared with previous methods, PRNet can calibrate $2\times $ of drifted sensors at the recovery rate of 80% under the same level of accuracy requirement. We also provide helpful insights for designing deep neural networks for sensor calibration. We hope our proposed simple and effective approach will serve as a solid baseline in blind drift calibration of sensor networks.

本文主要研究将无线倾角传感器应用于桥梁梁体挠度变化的长期监测.基于支座坐标与多个位置倾角监测数据,采用最小二乘算法构造了梁体下挠多项式曲线的拟合方法.在江苏南通如泰运河大桥的长期监测系统中部署了无线倾角传感器,获取了该桥的倾角监测数据,并根据3跨连续梁桥结构特点,选取6阶多项式进行拟舍得到全桥挠度曲线.通过相关分析验证了桥梁挠度与环境温度的相关性.本项目的实施说明了快速布设的无线倾角计可以实现低成本高可靠的长期挠度监测,在未来实际工程中可进行广泛应用.

The lifetime of wireless sensor networks (WSNs) has been significantly extended, while in long-term large-scale WSN applications, the increasing sensor drift has become a key problem affecting the reliability of sensory data. In this paper, we propose a blind online drift calibration framework based on subspace projection and sparse recovery for sensor networks in general-purpose monitoring. Temporal sparse Bayesian learning is used in the proposed method to estimate the sensor drift from under-sampled observations. The proposed method needs neither dense deployment nor the presence of a prior data model. Both simulated and real-world data set are used to evaluate the proposed method. Experimental results demonstrate that the proposed method can detect and recover the sensor drift when the number of drifted sensors are less than 20%, and when 40% sensors are drifted, the recovery rate is 80%.

针对桥梁结构监测采集到的桥梁异常状态下长期积累演变的惊人数据量,提出了基于主成分分析与人工神经网络相结合的桥梁结构异常状态识别方法.布设多种类型传感器监测获取高维数据,采用主成分分析法对原始高维特征数据进行预处理,将结构异常特征变量的主成分作为人工神经网络的输入特征.该方法有效的降低了神经网络的结构复杂度,同时提高人工神经网络的训练速度,也保证了人工神经网络具有良好的收敛性和稳定性.应用于江苏南通如泰运河大桥和江苏无锡开源大桥的实际监测数据的结果表明,基于主成分分析的人工神经网络方法用于桥梁结构异常状态识别,与传统的神经网络以及其他模式识别算法相比,有更好的识别精度.